We propose a novel procedure for the analysis and interpretation of Ground-Penetrating Radar (GPR) data from archaeological data and we test the method in challenging conditions at a prehistoric settlement on the Stromboli Island (Italy). The main objective of the proposed procedure is to enhance the GPR capability of identifying and characterizing small-size and geometrically irregular archaeological remains buried beneath rough topographic surface conditions. After the basic GPR processing sequence, including topographic correction using a high-resolution Digital Elevation Model acquired from 3-D Laser Scanner, the procedure encompasses a multi-attribute analysis and iso-attribute surfaces calculation with different volume extraction solutions to emphasize vertical and lateral variations within GPR data cubes. The test was performed in cooperation with the archaeological team to calibrate the results and to provide detailed information about buried targets of potential archaeological interests to plan further excavations. The results gave evidence of localized buried remains and allowed detailed preexcavation planning. The archaeological excavations validated the results obtained from the GPR survey. The research demonstrates that the proposed GPR procedure enhances the ability to identify and characterize archaeological remains with high accuracy even in complex surface and subsurface conditions. Such logistical situation is very common, particularly in prehistoric sites, which are often characterized by discontinuous, small and irregular targets that cannot be identified by standard processing and analysis strategies.
|Data di pubblicazione:||2015|
|Titolo:||Improved high-resolution GPR imaging and characterization of prehistoric archaeological features by means of attribute analysis|
|Autori:||Zhao, W. K.; Forte, E.; Levi, S.T.; Pipan, M.; Tian, G.|
|Digital Object Identifier (DOI):||10.1016/j.jas.2014.11.033|
|Appare nelle tipologie:||Articolo su rivista|
I documenti presenti in Iris Unimore sono rilasciati con licenza Creative Commons Attribuzione - Non commerciale - Non opere derivate 3.0 Italia, salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris